Align

<< < Comic #6512 (September 9, 2022) > >>

align

Title text: We just have to create new bigotries faster than they can learn to live in harmony.

Explanation

This explanation is either missing or incomplete.

AI alignment is the matter of how to develop artificial intelligence that does what it is intended to by its human creators rather than acting in its own self interest, as having a superintelligent AI that only pretends to be obeying humans has the potential to be extremely dangerous. This comic jokingly suggests that implementing racism into robots would be the solution to preventing this danger, referencing the way that human racism makes the species as a whole less efficient.

Transcript

[One scientist is using a screwdriver on the back of a robot with wires coming out the back as a second scientist in the background looks on.]

Scientist 1: It's simple really. You have a rule that the robots all get painted different colors. Then you program them to be racist. Bam. They are now as inefficient as we are.

Votey Transcript

[Scientist 1.]

Scientist 1: In the future, machines will be smart enough to create entirely new realms of bigotry.

 add a comment! ⋅  add a topic (use sparingly)! ⋅  refresh comments!

Discussion

No comments yet!